autoencoder

Who acts: kind, One, That, you, seasons, network · What happens: squeeze, grow back, learn, be, tell, press · To what, where: input, few numbers, copy, original, Middle, world · Speech: someone speaks here

«autoencoder» in the mysticism dictionary: its meaning for a pupil and its 5 examples, every sentence parsed by Mova itself — its syntax tree, a dependency tree in the style of Universal Dependencies, and its pragmatics clause by clause. The word in the dictionary, with its sense in the lessons: autoencoder in the mysticism dictionary.

Meaning 1

A kind of neural network that squeezes its input into a few numbers and then grows it back, and learns from how far the copy is from the original.

The tree, 29 words

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	A	a	DET	Definite=Ind PronType=Art	2 kind	det
2	kind	kind	NOUN	Number=Sing	30 original	nsubj
3	of	of	ADP	_	5 network	case
4	neural	neural	ADJ	Degree=Pos	5 network	amod
5	network	network	NOUN	Number=Sing	2 kind	nmod
6	that	that	PRON	PronType=Rel	7 squeezes	nsubj
7	squeezes	squeeze	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	2 kind	acl:relcl
8	its	its	PRON	Case=Gen Gender=Neut Number=Sing Person=3 Poss=Yes PronType=Prs	9 input	nmod:poss
9	input	input	NOUN	Number=Sing	7 squeezes	obj
10	into	into	ADP	_	13 numbers	case
11	a	a	DET	Definite=Ind PronType=Art	13 numbers	det
12	few	few	ADJ	Degree=Pos	13 numbers	amod
13	numbers	number	NOUN	Number=Plur	7 squeezes	obl
14	and	and	CCONJ	_	16 grows	cc
15	then	then	ADV	PronType=Dem	16 grows	advmod
16	grows	grow	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	7 squeezes	conj
17	it	it	PRON	Case=Acc Gender=Neut Number=Sing Person=3 PronType=Prs	16 grows	obj
18	back	back	ADV	_	16 grows	advmod
19	,	,	PUNCT	_	21 learns	punct
20	and	and	CCONJ	_	21 learns	cc
21	learns	learn	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	7 squeezes	conj
22	from	from	ADP	_	26 copy	case
23	how	how	ADV	PronType=Int	24 far	advmod
24	far	far	ADV	Degree=Pos	26 copy	advmod
25	the	the	DET	Definite=Def PronType=Art	26 copy	det
26	copy	copy	NOUN	Number=Sing	21 learns	obl
27	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	30 original	cop
28	from	from	ADP	_	30 original	case
29	the	the	DET	Definite=Def PronType=Art	30 original	det
30	original	original	NOUN	Number=Sing	root	root
31	.	.	PUNCT	_	30 original	punct

#	verb	tense, mood	who	what, whom	where	when	roles by
1	squeeze	present	a kind of neural network	kind's input	a few numbers		vn knead-26.5 «NP V NP.material PP.product»: agent=Agent input=Material into number=Product
2	grow back	present	a kind of neural network not said here	input		then	vn appear-48.1.1 «PP.location V NP»: agent=Location it=Theme
3	learn	present	a kind of neural network not said here		the copy		vn learn-14-2-1 «NP V PP.source»: agent=Agent from copy=Source
4	is	present	a kind of neural network		the original		tree rules

One row per clause the reader finds. «not said here»: a role filled from the clause above. «roles by»: a VerbNet class and frame when one fits the verb, else the reader's rules on the tree.

Example 1

The One in the Middle told it as an autoencoder: a world pressed into a point, and unfolded again. (note 05)

Sentence 1: The One in the Middle told it as an autoencoder: a world pressed into a point, and unfolded again.

The tree, 19 words

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	The	the	DET	Definite=Def PronType=Art	2 One	det
2	One	one	NUM	NumForm=Word NumType=Card	6 told	nsubj
3	in	in	ADP	_	5 Middle	case
4	the	the	DET	Definite=Def PronType=Art	5 Middle	det
5	Middle	Middle	PROPN	Number=Sing	6 told	obl
6	told	tell	VERB	Mood=Ind Number=Plur Person=3 Tense=Past VerbForm=Fin	root	root
7	it	it	PRON	Case=Acc Gender=Neut Number=Sing Person=3 PronType=Prs	6 told	obj
8	as	as	ADP	_	10 autoencoder	case
9	an	a	DET	Definite=Ind PronType=Art	10 autoencoder	det
10	autoencoder	autoencoder	NOUN	Number=Sing	6 told	obl
11	:	:	PUNCT	_	13 world	punct
12	a	a	DET	Definite=Ind PronType=Art	13 world	det
13	world	world	NOUN	Number=Sing	10 autoencoder	appos
14	pressed	press	VERB	Tense=Past VerbForm=Part Voice=Pass	13 world	acl
15	into	into	ADP	_	17 point	case
16	a	a	DET	Definite=Ind PronType=Art	17 point	det
17	point	point	NOUN	Number=Sing	14 pressed	obl
18	,	,	PUNCT	_	20 unfolded	punct
19	and	and	CCONJ	_	20 unfolded	cc
20	unfolded	unfold	VERB	Tense=Past VerbForm=Part Voice=Pass	6 told	conj
21	again	again	ADV	_	20 unfolded	advmod
22	.	.	PUNCT	_	6 told	punct

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	tell speech: assert	past	the One		Middle	the Middle		as a autoencoder	vn transfer_mesg-37.1.1-1-1 «NP V NP.recipient»: agent=Agent it=Recipient
2	press	participle		a world not said here		a point			vn push-12-1-1 «NP V PP.trajectory»: agent=Agent into point=Trajectory
3	unfold	participle	the One not said here				again		vn bend-45.2 «NP.patient V»: agent=Patient

Sentence 2: (note 05)

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	(	(	PUNCT	_	2 note	punct
2	note	note	NOUN	Number=Sing	root	root
3	05	05	NUM	NumForm=Digit NumType=Card	2 note	flat
4	)	)	PUNCT	_	2 note	punct

The frames reader finds no event or fact in this sentence.

Example 2

That is the shape of an autoencoder, read backwards: the seasons decode first and encode after, where a network encodes first. (note 08)

Sentence 1: That is the shape of an autoencoder, read backwards: the seasons decode first and encode after, where a network encodes first.

The tree, 21 words

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	That	that	PRON	Number=Sing PronType=Dem	4 shape	nsubj
2	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	4 shape	cop
3	the	the	DET	Definite=Def PronType=Art	4 shape	det
4	shape	shape	NOUN	Number=Sing	9 read	obj
5	of	of	ADP	_	7 autoencoder	case
6	an	a	DET	Definite=Ind PronType=Art	7 autoencoder	det
7	autoencoder	autoencoder	NOUN	Number=Sing	4 shape	nmod
8	,	,	PUNCT	_	4 shape	punct
9	read	read	VERB	Mood=Imp VerbForm=Fin	root	root
10	backwards	backwards	ADV	_	9 read	advmod
11	:	:	PUNCT	_	14 decode	punct
12	the	the	DET	Definite=Def PronType=Art	13 seasons	det
13	seasons	season	NOUN	Number=Plur	14 decode	nsubj
14	decode	decode	VERB	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	9 read	parataxis
15	first	first	ADV	NumForm=Word NumType=Ord	14 decode	advmod
16	and	and	CCONJ	_	17 encode	cc
17	encode	encode	VERB	VerbForm=Inf	14 decode	conj
18	after	after	ADV	_	17 encode	advmod
19	,	,	PUNCT	_	23 encodes	punct
20	where	where	ADV	PronType=Rel	23 encodes	advmod
21	a	a	DET	Definite=Ind PronType=Art	22 network	det
22	network	network	NOUN	Number=Sing	23 encodes	nsubj
23	encodes	encode	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	17 encode	conj
24	first	first	ADV	NumForm=Word NumType=Ord	23 encodes	advmod
25	.	.	PUNCT	_	9 read	punct

#	verb	tense, mood	who	what, whom	other	roles by
1	is a	present	That	the shape	of a autoencoder	tree rules
2	read	imperative	you not said here		backwards property	vn transfer_mesg-37.1.1-1 «NP V»: agent=Agent
3	decode	present	the seasons		first property	tree rules
4	encode	infinitive	the seasons not said here		after property	tree rules
5	encode	present	a network		first property	tree rules

Sentence 2: (note 08)

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	(	(	PUNCT	_	2 note	punct
2	note	note	VERB	Mood=Imp VerbForm=Fin	root	root
3	08	08	NUM	NumForm=Digit NumType=Card	2 note	obj
4	)	)	PUNCT	_	2 note	punct

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	note	imperative	you not said here	08					vn say-37.7-1-1 «NP V NP.topic»: agent=Agent 08=Topic

Example 3

An autoencoder is trained on its reconstruction error: how far the world that comes out is from the world that went in. (note 08)

Sentence 1: An autoencoder is trained on its reconstruction error: how far the world that comes out is from the world that went in.

The tree, 22 words

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	An	a	DET	Definite=Ind PronType=Art	2 autoencoder	det
2	autoencoder	autoencoder	NOUN	Number=Sing	4 trained	nsubj:pass
3	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	4 trained	aux:pass
4	trained	train	VERB	Tense=Past VerbForm=Part Voice=Pass	root	root
5	on	on	ADP	_	8 error	case
6	its	its	PRON	Case=Gen Gender=Neut Number=Sing Person=3 Poss=Yes PronType=Prs	8 error	nmod:poss
7	reconstruction	reconstruction	NOUN	Number=Sing	8 error	compound
8	error	error	NOUN	Number=Sing	4 trained	obl
9	:	:	PUNCT	_	20 world	punct
10	how	how	ADV	PronType=Int	11 far	advmod
11	far	far	ADV	Degree=Pos	20 world	advmod
12	the	the	DET	Definite=Def PronType=Art	13 world	det
13	world	world	NOUN	Number=Sing	20 world	nsubj
14	that	that	PRON	PronType=Rel	15 comes	nsubj
15	comes	come	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	13 world	acl:relcl
16	out	out	ADP	_	15 comes	compound:prt
17	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	20 world	cop
18	from	from	ADP	_	20 world	case
19	the	the	DET	Definite=Def PronType=Art	20 world	det
20	world	world	NOUN	Number=Sing	4 trained	obl
21	that	that	PRON	PronType=Rel	22 went	nsubj
22	went	go	VERB	Mood=Ind Number=Sing Person=3 Tense=Past VerbForm=Fin	20 world	acl:relcl
23	in	in	ADV	_	22 went	advmod
24	.	.	PUNCT	_	4 trained	punct

#	verb	tense, mood	who	what, whom	where	other	roles by
1	train	present · passive		a autoencoder	autoencoder's reconstruction error		vn hire-13.5.3 «NP V NP»: subject=Theme
2	come out	present	the world				vn become-109.1 «NP V ADJ»: agent=Patient
3	is	present	the world		the world	far property	tree rules
4	go in	past	the world				vn disappearance-48.2-1 «NP V»: agent=Theme
5	err		autoencoder				nom error «act intransitive»: poss=agent

Sentence 2: (note 08)

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	(	(	PUNCT	_	2 note	punct
2	note	note	VERB	Mood=Imp VerbForm=Fin	root	root
3	08	08	NUM	NumForm=Digit NumType=Card	2 note	obj
4	)	)	PUNCT	_	2 note	punct

#	verb	tense, mood	who	what, whom	to whom	where	when	other	roles by
1	note	imperative	you not said here	08					vn say-37.7-1-1 «NP V NP.topic»: agent=Agent 08=Topic

Example 4

That is the shape of an autoencoder: layers widen, narrow to a bottleneck, widen again; the seventh day is the latent space, where the whole of creation fits into one thought, and Mova does the same in miniature, text into a graph and back. (lesson 020)

Sentence 1: That is the shape of an autoencoder: layers widen, narrow to a bottleneck, widen again; the seventh day is the latent space, where the whole of creation fits into one thought, and Mova does the same in miniature, text into a graph and back.

The tree, 44 words

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	That	that	PRON	Number=Sing PronType=Dem	4 shape	nsubj
2	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	4 shape	cop
3	the	the	DET	Definite=Def PronType=Art	4 shape	det
4	shape	shape	NOUN	Number=Sing	root	root
5	of	of	ADP	_	7 autoencoder	case
6	an	a	DET	Definite=Ind PronType=Art	7 autoencoder	det
7	autoencoder	autoencoder	NOUN	Number=Sing	4 shape	nmod
8	:	:	PUNCT	_	10 widen	punct
9	layers	layer	NOUN	Number=Plur	10 widen	nsubj
10	widen	widen	VERB	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	4 shape	parataxis
11	,	,	PUNCT	_	12 narrow	punct
12	narrow	narrow	ADJ	Degree=Pos	10 widen	advcl
13	to	to	ADP	_	15 bottleneck	case
14	a	a	DET	Definite=Ind PronType=Art	15 bottleneck	det
15	bottleneck	bottleneck	NOUN	Number=Sing	12 narrow	obl
16	,	,	PUNCT	_	17 widen	punct
17	widen	widen	VERB	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	10 widen	conj
18	again	again	ADV	_	17 widen	advmod
19	;	;	PUNCT	_	26 space	punct
20	the	the	DET	Definite=Def PronType=Art	22 day	det
21	seventh	seventh	ADJ	Degree=Pos	22 day	amod
22	day	day	NOUN	Number=Sing	26 space	nsubj
23	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	26 space	cop
24	the	the	DET	Definite=Def PronType=Art	26 space	det
25	latent	latent	ADJ	Degree=Pos	26 space	amod
26	space	space	NOUN	Number=Sing	33 fits	nsubj
27	,	,	PUNCT	_	30 whole	punct
28	where	where	ADV	PronType=Rel	30 whole	advmod
29	the	the	DET	Definite=Def PronType=Art	30 whole	det
30	whole	whole	ADJ	Degree=Pos	26 space	amod
31	of	of	ADP	_	32 creation	case
32	creation	creation	NOUN	Number=Sing	30 whole	obl
33	fits	fit	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	4 shape	parataxis
34	into	into	ADP	_	36 thought	case
35	one	one	NUM	NumForm=Word NumType=Card	36 thought	nummod
36	thought	thought	NOUN	Number=Sing	33 fits	obl
37	,	,	PUNCT	_	40 does	punct
38	and	and	CCONJ	_	40 does	cc
39	Mova	Mova	PROPN	Number=Sing	40 does	nsubj
40	does	do	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	33 fits	conj
41	the	the	DET	Definite=Def PronType=Art	42 same	det
42	same	same	ADJ	Degree=Pos	40 does	obj
43	in	in	ADP	_	44 miniature	case
44	miniature	miniature	NOUN	Number=Sing	40 does	obl
45	,	,	PUNCT	_	46 text	punct
46	text	text	NOUN	Number=Sing	44 miniature	conj
47	into	into	ADP	_	49 graph	case
48	a	a	DET	Definite=Ind PronType=Art	49 graph	det
49	graph	graph	NOUN	Number=Sing	40 does	obl
50	and	and	CCONJ	_	51 back	cc
51	back	back	ADV	_	49 graph	conj
52	.	.	PUNCT	_	4 shape	punct

#	verb	tense, mood	who	what, whom	where	when	other	roles by
1	is a	present	That	the shape			of a autoencoder	tree rules
2	widen	present	layers					vn other_cos-45.4 «NP.patient V»: agent=Patient
3	widen	present	layers not said here			again		vn other_cos-45.4 «NP.patient V»: agent=Patient
4	is a	present	the seventh day	the latent space				tree rules
5	fit	present	the latent space		one thought			vn funnel-9.3-1 «NP V NP PP.destination»: agent=Agent into thought=Destination
6	do	present	Mova	the same	miniature and text		into a graph	vn act-114-1 «NP V NP»: agent=Agent same=Eventuality

Sentence 2: (lesson 020)

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	(	(	PUNCT	_	2 lesson	punct
2	lesson	lesson	NOUN	Number=Sing	root	root
3	020	020	NUM	NumForm=Digit NumType=Card	2 lesson	flat
4	)	)	PUNCT	_	2 lesson	punct

The frames reader finds no event or fact in this sentence.

Example 5

The note says this is the shape of an autoencoder: the layers widen, narrow to a bottleneck, and widen again. (lesson 020)

Sentence 1: The note says this is the shape of an autoencoder: the layers widen, narrow to a bottleneck, and widen again.

The tree, 20 words

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	The	the	DET	Definite=Def PronType=Art	2 note	det
2	note	note	NOUN	Number=Sing	3 says	nsubj
3	says	say	VERB	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	root	root
4	this	this	PRON	Number=Sing PronType=Dem	7 shape	nsubj
5	is	be	AUX	Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin	7 shape	cop
6	the	the	DET	Definite=Def PronType=Art	7 shape	det
7	shape	shape	NOUN	Number=Sing	3 says	ccomp
8	of	of	ADP	_	10 autoencoder	case
9	an	a	DET	Definite=Ind PronType=Art	10 autoencoder	det
10	autoencoder	autoencoder	NOUN	Number=Sing	7 shape	nmod
11	:	:	PUNCT	_	14 widen	punct
12	the	the	DET	Definite=Def PronType=Art	13 layers	det
13	layers	layer	NOUN	Number=Plur	14 widen	nsubj
14	widen	widen	VERB	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	3 says	parataxis
15	,	,	PUNCT	_	16 narrow	punct
16	narrow	narrow	ADJ	Degree=Pos	14 widen	advcl
17	to	to	ADP	_	19 bottleneck	case
18	a	a	DET	Definite=Ind PronType=Art	19 bottleneck	det
19	bottleneck	bottleneck	NOUN	Number=Sing	16 narrow	obl
20	,	,	PUNCT	_	22 widen	punct
21	and	and	CCONJ	_	22 widen	cc
22	widen	widen	VERB	Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin	14 widen	conj
23	again	again	ADV	_	22 widen	advmod
24	.	.	PUNCT	_	3 says	punct

#	verb	tense, mood	who	what, whom	when	other	roles by
1	say speech: assert	present	the note				vn say-37.7-1-1 «NP V that S»: agent=Agent
2	is a	present	this	the shape		of a autoencoder	tree rules
3	widen	present	the layers				vn other_cos-45.4 «NP.patient V»: agent=Patient
4	widen	present	the layers not said here		again		vn other_cos-45.4 «NP.patient V»: agent=Patient

Sentence 2: (lesson 020)

The words, one row each (CoNLL-U)

#	word	lemma	UPOS	features	head	relation
1	(	(	PUNCT	_	2 lesson	punct
2	lesson	lesson	NOUN	Number=Sing	root	root
3	020	020	NUM	NumForm=Digit NumType=Card	2 lesson	flat
4	)	)	PUNCT	_	2 lesson	punct

The frames reader finds no event or fact in this sentence.

Text: the mysticism dictionary of Mova's schools (world/data/dictionary-schools/), written for Mova. Trees: Mova's own parse — the en model on the CPU, when the site is built — not checked by a teacher; they may be wrong. Frames: Mova's frames reader (prag), each meaning and example read on its own. Elamur Cosmic License.

This page in Markdown → https://md.mova.elamur.ai/free/en/corpora/dictionaries/mysticism/A/autoencoder/